ix the tropics : rn. 101 



METHODS EMPLOYED. 



The plants were grown as a rule in rows 3 feet apart, the 

 distance between individual grains sown being 1 foot in the 

 case of the larger varieties, and rather less in that of the smaller. 



In the first instance a number of direct pollinations were 

 effected between particular parent plants, the method being 

 to enclose the stigmas from the time of their first appearance 

 in a bag of parchment paper, the male inflorescences being- 

 treated in the same way, and then to transfer the whole m;is< 

 of male flowers to the bag enclosing the stigmas of the plant 

 to be pollinated. Even by this method it is not possible to be 

 certain that foreign pollen is completely excluded, and the 

 result of some of the first experiments with the blue character 

 led me to suppose the method to be much less accurate than 

 was really the case. 



Afterwards the following method was adopted. The grains 

 of the recessive strain which was to be used as the pollen parent 

 were sown in alternate rows upon an isolated plot of ground of 

 considerable size, and in the remaining rows were planted the 

 various forms which it was desired to test by crossing with 

 the recessive. The tassels of the seed parents were removed 

 previous to the opening of the staminal flowers, and pollinat i. m 

 was left to the agency of the wind. 



This method is open to certain objections, the most serious 

 of which lies in the fact that the individual pollinating 

 parentis not known in any particular case, so that possible 

 individual differences between the members of the recessive 

 strain used are quite uncontrolled. Even for a broadly statis 

 tical result it is necessary that a very uniform strain should 

 be used as the source of pollen, and the point must be 

 tested by examining the cobs produced by every plant of the 

 pollinating form. If a single wrong seed escape notice in the 

 sowing, the plant arising from it may wrongly pollinate all its 

 neighbours. Great care must further be exercised in removing 

 all the staminal flowers of the seed parents before any pollen 

 escapes from them ; and some are very Liable t<> he overlooked 



